April 26, 2022

Introduction

Our Questions

  • Do COVID case numbers affect Zoom lecture attendance?
  • Does weather affect Zoom lecture attendance?

Gathering Data

Operationalization

  • Weather
    • type (☀️, ☁️, ☔, ❄️, etc.)
    • temperature (T-2, T-1, T-0)
  • COVID ( Univ. of Utah, Salt Lake County)
    • Daily case numbers
    • 7-day average

Collection: Scraper

Collection: Zoom Attendance

Collection: Zoom Attendance

  • Zoom attendence recorded via Zoom headcount
  • Recorded at
    • 10 min after start
    • Half-way
    • 10 min before end
  • Minus one for professor
  • Minus one for researcher (us)
  • Left TAs in the count

Collection: University of Utah Covid Data

  • Available on U of U Covid infomration page
  • Total new cases and 7-day avearge new cases
    • Among faculty, staff, and students
  • U of U stopped posting data on March 3rd
    • Couldn’t get data retroactively
  • Scraper used to collect data

Collection: Salt Lake County Covid Data

  • COVID-19 data downloaded from the COVID Central
    • Daily new case counts
    • 7-day average new cases
    • 7-day average percent positive test rate
  • Used Salt Lake County entries

Weather Data

  • Collected from OpenWeatherMap
  • Weather type (☀️, ☁️, ☔, ❄️, etc.)
    • Collected 2hrs before class started
  • Temperature
    • Collected 0hrs, 1hr, and 2hrs before class
    • Average daily temperature
  • Used scraper to collect data

Plots! (1/3)

  • Derek’s plots here

Plots! (2/3)

  • More of Derek’s plots

Plots! (3/3)

  • More of Derek’s plots? etc.

Analysis (1/3)

Data

  • Zoom attendance
  • Daily new case counts SL County
  • 7-day average new case count SL County
  • 7-day average positive test rate SL County
  • Average daily temperature SL City

Analysis (2/3)

  • Mean Daily Attendance Rate
    • Aggregated course Zoom attendance into single attendance rate for each day
  • Pearson correlation coefficient calculated
  • Fisher transformation applied to yield a normally distributed variable
  • 95% confidence interval calculated for each correlation coefficient

Analysis Results? (3/3)

  • Mean Atd Rate vs Mean Daily Temperature
    • CC: -0.0392 & CI: -0.3725, 0.3029
  • Mean Atd Rate vs Daily Confirmed New Case Count in SL County
    • CC: -0.1932 & CI: -0.4987, 0.1551
  • Mean Atd Rate vs 7-Day Average Confirmed New Case Count in SL County
    • CC: -0.1064 & CI: -0.4291, 0.2404
  • Mean Atd Rate vs 7-Day Average Positive Test Rate in SL County
    • CC: -0.0472 & CI: -0.3793, 0.2957

No correlation? Inconclusive?

Analysis (2/?)

A different approach. 🤠 Data wrangling:

data <- read.csv("data/atd_weather_full.csv") %>%
    rowwise() %>%
    mutate(mean_atd = mean(c(atd_start, atd_mid, atd_end))) %>%
    mutate(atd_rate = mean_atd / enrolled) %>%
    mutate(course = factor(course)) %>%
    mutate(weather = factor(weather)) %>%
    mutate(date = as.Date(date, "%m/%d/%Y")) %>%
    mutate(mean_temp = mean(temp_tm2, temp_tm1, temp_tm0, na.rm = TRUE))

First, Zoom attendance rate versus mean temp:

Zoom attendance rate versus the 7-day average of new COVID cases at the :

So, does this mean me might see relationships within individual classes?

Let’s find out!

Analysis (?/?)

3200, 3500, and 4400 look promising: we’ll make a plot matrix:

p3500 <- data_filtered %>%
    filter(course == 3500) %>%
    select(!course) %>%
    ggpairs() +
    theme(axis.text.x = element_text(angle = 90, hjust = 1),
          axis.title = element_text(size = rel(0.85))) +
    labs(title = "3500 Pairs")

(repeat as needed)

Conclusions! …maybe?

  • Zoom Attendance Rate vs. Mean Temp
    • CS 3500: (\(p > 0.05\)) – Not significant.
    • CS 3200: \(-0.564,\ R^2 = 0.318\) (\(p < 0.01\))
    • CS 4400: \(0.6,\ R^2 = 0.36\) (\(p < 0.01\))

  • Zoom Attendance Rate vs. 7-day Avg Cases at
    • CS 3500: \(-0.124,\ R^2 = 0.015\) (\(p < 0.001\))!!
    • CS 3200: \(-0.564,\ R^2 = 0.318\) (\(p < 0.01\))
    • CS 4400: (\(p > 0.05\)) – Not significant.

Conclusions? …perhaps not.

  • Limitations too great to make assertions…